There are two main pharmacotherapy agents used in the maintenance treatment of opioid addicts: the narcotic agonist methadone and the narcotic antagonist, naltrexone. Each treatment has limitations: for naltrexone, poor patient retention, and for methadone maintenance, continued illicit opioid use in patients maintained on relatively low methadone doses and poor success at detoxification. Given these difficulties with each maintenance treatment, this study was designed to evaluate a new maintenance agent - buprenorphine, a mixed agonist/antagonist. Unlike other opioid agonists, buprenorphine does not lead to significant physical dependence in man and appears to prevent narcotic euphoria and physical dependence through both cross tolerance and competitive antagonism. This study compares the efficacy of buprenorphine maintenance (4 mg daily with possible increase to 8 mg) to methadone maintenance (35 mg daily with possible increase to 65 mg) in a 6-month outpatient maintenance program using a double blind, random assignment trial. One hundred and twenty (120) patients will be used to assess efficacy, as defined relative to treatment retention, illicit drug use, and symptom levels (withdrawal, depression, and anxiety) during the 6-month trial. Data analyses will involve life table analysis of length of time in treatment and time remaining abstinent from illicit drugs, as well as analysis of variance for repeated measures on the withdrawal symptom, depression and anxiety scores comparing the buprenorphine to the methadone maintained group. We hypothesize that although buprenorphine may be no better than methadone in retaining patients in treatment, it will be superior in rates of abstinence from illicit drugs during the 6 months of maintenance and/or in successful detoxification to a medication-free state during the 6 month follow-up.